CLASSIF1 Data Pattern Classificaition


Patient Prognosis after Melanoma Surgery

Reclassification of the Learning Set




< fig.1      fig.2: Reclassification of the Learning Set      > fig.3

- Displayed are the triple matrix patterns of the first 10 patients of the survivor (S) and non survivor (N) melanoma patients after surgery.

- The disease classification mask for survivors contains typically a sequence of (0) characters because the majority of the parameter values is located between the two percentiles thresholds for the percentile ranges 10-90%, 15-85%, 20-80%, 25-75%, 30-70%. For the percentile pair 10-90%, 80% of the values are (0), 10% below (-) the lower percentile and 10% above (+) the upper percentile while the numbers are 40%, 30%, 30% in case of the 30-70% percentile pair.

- Unknown patients are classified according to the highest positional coincidence of the patient classification mask with any of the disease classification masks.

- The degree of coincidence between the patient mask and the best fitting disease mask is expressed by the mask coincidence factor. The coincidence factors are 1.00 for survivors #009, 012 despite the fact that not all triple matrix characters are (0) (arrows <= ). Only survivors #002, 006, 008 show complete coincidence between their patient classification mask and the all (0) disease classification mask for survivors. Non survivors have on average all discriminatory parameters of the disease classification mask increased (+). For this reason, patients showing a majority of normal (0) or diminished (-) parametervalues are attributed to the survivor category, explaining the coincidence factor of 1.00 for patients #009, 012.


© 2026 G.Valet
1965-2006: Max-Planck-Institut für Biochemie, Am Klopferspitz 18a, D-82152 Martinsried, Germany
Last update: Apr 18,2002
First display: Mar 21,2002